A Proposal for the Sciences, Rediscover Bacon's Idols of the Mind

Why is science so stubbornly politicized?  Perhaps we have just been teaching it wrong.

Francis Bacon benefited from his Queen's loyalty to the split with the Catholic church. Being entirely separate had its advantages, and Bacon's firm adherence to Aristotle while despising the state of received Aristotelian philosophy materialized in his bold new vision for the scientific approach.

I am happy to discuss in comments the details of this matter, which I am still working through, but let's make a long story short. Science before Bacon had been captured by the idea of the deductive method. This means that first principles were hypothesized which, when logically applied, would sort out almost any contradiction between competing claims. What if these didn't work because of some new discovery? Simply, revise the first principles. This process would give us insight into the true nature of the world.

Bacon offered an alternative. His method looked towards the inductive method. This takes a radical approach to science that we are all taught in our first laboratory class, the scientific method. Simply put, the scientific method asks us which phenomenon we are interested in and then accumulates related observations. We sort these observations by their perceived priority in association with the main phenomenon of interest.

In this approach, Bacon discovered what he called the Idols of the Mind. He gave them names:

• idola tribus: The idol of the "tribe" deals with coherence of ideas into neat packages. I take this to be related to much of the moral foundations literature Jonathan Haidt is working with. 
• idola spectus: The idol of the "cave" deals with cognitive bias. For example, our minds interpret our enemies best argument as evidence that they have bad intentions. Liberals who watch Fox News confirm their prior that conservatives are crazy. 
• idola fori: The idol of the "marketplace"  deals with language. Often language assumes a meaning quite different the way specialist use it, this helps distort careful logic when applied to the public debate.
• idola theatri: The idol of the "theatre" is the closes to sophistry. The tendency to over-sell academic expertise. 
• idola schola: The idol of the "school" is the pedantic reasoning of textbooks. It is a oversimplification useful only for the teaching of material to neophytes. 

The idols were categories of errors. Science according to Bacon was understanding error, not finding truth.  The most important lesson that Bacon teaches us is that all answers are errors and that these are more or less affected by these idols. To begin the process of cleansing the mind from these idols, there are many approaches. None are going to work. We can only progress by taking the task seriously. 

I find it discouraging sometimes that so many students think that they can posses "the answer" and have simply not been taught the idols of the mind, specifically idola schola. I can't read a periodical without thinking of the idola fori. I have so much contempt for my peers who dismiss the idola theatri and attempt to establish their reputation by sophism. 

The only hope science has is to proceed with a more ready understanding of Bacon's idols. It strikes me that much of what we see as progress in the last 300 years depends on the implicit expectation of this task. If anything explains why modern science is so politicized, it is the fact that we have lost our understanding of Bacon's method. I aim to do better.    

More on Rules and a little on Bentham

Diana starts off the latest round of conversation talking about rules. I want to even take it back to a more basic point about rules.

What are rules designed to do?

If we proceed from a rule-utilitarian perspective than we have two routes to follow that I can think of. John Rawls has broken the path so well in terms of his reflexive equilibrium that many may only think of his approach, but I also want to consider a different approach that is less obvious. But first, we focus on Rawls.

The reflexive equilibrium is designed to produce rules that coincide with a desirable resulting state-of-the-world. Rules are adjusted like the throttle of a motor-bike to achieve the desired result. Each throttle is tuned to produce incentives necessary to reach the desired outcome. However, Rawls is not so naive as to not give us guidance as to what these throttles are aiming to achieve (his version of the teleos). The primary purpose is to achieve fairness and the possibility of allowing unfairness is only allowed in a dynamic context if the ultimate growth which results form the difference increases the least-well-off. Well certainly this changes the game for interpreting the rule-utilitarian approach. I defy you to find a better discussion of rule-utilitarianism than that which is surrounding the work of John Rawls. Even in economics, James Buchanan relies heavily on Rawls (both took Frank Knight very seriously, BTW). All of this is very rational and very optimistic about human reason. It also enshrines stability and order in rules.

Now that I have told you want is the standard approach to rule-utilitarianism, I want to go further. What I have discovered in reading late 18th and early 19th century texts is that before it was clarified by Rawls, some other thoughts were out there. The name of the game was to replace the interpretation of morality by clerics and create a secular basis for rationally articulating moral reasoning.

The work of the arch-utilitarian Jeremy Bentham is dripping with moral imagery while simultaneously dismissing rights-theory and organized religion. Bentham rejects what Hume would call superstition and enthusiasm, but he holds onto the subjective. [This last line is my own reading]. So what does a utilitarianism look like that includes this type of pursuit. It is something that is larger than the utilitarianism you have been taught was in Bentham.

For example: Bentham considers the question of committing suicide publicly when one's spouse dies. He considers the utility calculation for the individual to justify the position and to argue against it. Certainly the actor in question could be making an accurate representation of their decision, but also the person could feel as if society has put a great deal of pressure on them to do this. Barring an environment where we were clear about the person's actual intentions we would be against such a practice. We could assume society is implicitly compelling the act. However, were we to be convinced of the sincerity of their motivation to die with their spouse, we would be forced to stand aside and allow it to take place.

For a more pedestrian example, should people who borrow money be allowed to do so at high rates? They certainly are higher risk, but who if not the market should decide about the appropriate level of risk in a society (by bidding up returns)? Too many risk-lovers and the market becomes brittle. The 1997 Asian crisis is one concrete example of excessively brittle international relations. The US banking disaster in 2008 is another example of high risk. So what is a rule-utilitarian to decide? Why is it that Bentham breaks from other rule-utilitarians and critiques financial regulation (famously arguing that Smith's defense of usury laws was inconsistent with other parts of his own work)? Bentham believes that society progresses when people take risks. He also longs for more progress which he defines as a result of risk-taking. To use rule-utilitarianism as it is often used, to constrain growth in favor of stability, is to bias the result. There are types of rule-utilitarianism that favor risk, for example.

Ultimately, I want to construct an analysis of rule-utilitarians that distinguishes them according to their time-preference or preference for the status quo vs. progress. The results fly in the face of conventional left-right dynamics. For instance, social democrats and their penchant for redistribution implicitly believe in lower growth rates than do right wingers who are willing to embrace social mobility and high growth. I argue that the conservative / progressive dichotomy is reversed at least on this margin.

What say you?

The Three "P"s and the Three "I"s

I channeled my dissertation advisor today and started to teach his rule, you can't get the three "I"s without the three "P"s.

This means that (P)roperty rights, (P)rice mechanisms, and (P)rofit and loss signals are all necessary to get (I)ncentives, (I)nformation, and (I)nnovation.

For the Incentives we have a large body of economic thought which develops the conflict between pro-market and anti-market types over the role that property rights play in aligning incentives.

The basic or first-approximation approach, I argued, was to simply point out that there was free-riding in the absence of well-defined property enforcement regimes. If you simply say "to each according to their need and from each according to their ability," this will end up with little accomplished in groups larger than a family. The next step is to teach the tragedy of the commons. The result of this story is that common property does not work well. Taken literally, the commons could be sold as private property and much of the problem would go away. So far the property rights story seems to be pretty solid.

There is a breakdown when we get to the story of externalities. Because externalities indicate places where property rights are ambiguous and are difficult to enforce, we don't necessarily think that property rights would be easy to define. In the case of planting flowers in your front yard, excluding those that refuse to contribute to their upkeep might be prohibitively difficult.

So, this leads us to a debate, one that is complex and difficult. Without the informational component of prices the property question would only move the debate to the abstract and theoretical level. The beauty of Mises and Hayek is that they make the informational argument so central to their discussion that it does not require the removal of all ambiguity about assigning property rights.

The only limitation that I can think of to the informational argument is that some things cannot be priced. I don't put a price (at least a money price) on love. There is something in us that resists the application of strict measures of scarcity to love. Now, this doesn't mean that we can't compare the marginal benefit another hour with our chosen special one gives us with the marginal benefit from something like writing a blog post. What it does say is that these calculations are going to be more difficult when the market is not thick and the transactions are not cleanly observable. Who is to actually price my preferences but me?

The final leg of this discussion is in Schumpeter. His creative destruction relies on change and it is no accident that Hayek in "The Use of Knowledge in Society"ends with Schumpeter (even if he is critiquing him). Schumpeter allows for change as the entrepreneur recognizes the opportunities to put innovation into effect. Schumpeter distinguishes between innovation and invention because invention doesn't do any good until it is applied to some purpose. The very possibility of change makes economics necessary (otherwise we would only solve the economic problem once and be done).

The link of these three "I"s to the Three "P"s is strong. Prices generate information, but only in the presence of profit and loss. Property rights give incentives and help prices to form better information. Innovation is not possible without some picture of the world created by the entrepreneur who has local knowledge.

Thank you Peter B.

Schumpeter and Innovation

Schumpeter is famous for creative destruction. However, what happens when the destruction is emphasized more than the creativity? What if the creativity is not immediately obvious?

Did anyone know that Velcro would be the next show fastener when it was created? I doubt it. Some discoveries are made by mistake. Most searches for particular answers take you down paths turning out far different than you would have thought. Many prototypes are far different than their mass produced grandchildren.

For these reasons and more the existence of creativity might be hidden, but the existence of destruction is highly visible to those that experience it. For many romantic reasons we think of people that have been doing something a certain way for generations as unimpeachable. However, there is asymmetry toward those that come up with a more efficient and exciting way to do something. Is this because early adopters are just really bad at sharing their enthusiasm over the new products?

In grad school we used to say all the time: "The market loves you," every-time some new consumer surplus was identified. This also manifest itself in Deidre McCloskey's line: "Why are there no folk songs for capitalism?"

Schumpeter's thesis in Capitalism, Socialism, and Democracy can be summarized in the following way: Over time people focus less on the creative and more on the destructive. This suggest something like constant attention paid to the misery that is created by changes over time while the possibilities that are opened up are appreciated at decreasing rates over time. If we plot these, we get a equilibrium where the line crosses the decelerating function of creativity (like the Solow growth model, for those in the know).

This means that unless creativity is actively appreciated, it can be lost in the complaints about how so many are being left out of the advantages of the modern world. Nassau Senior suggested that the luxuries of today become the decencies of tomorrow and the decencies of today become the necessities of tomorrow. This theory suggested that people's expectations change over time about what is necessary. Today's necessities include a color TV, for even the most poor people in the US have this item, considered a luxury 50 years ago. The case for access to medicine is the most striking. Hardly anyone could have heart surgery 50 years ago. Now it is often covered for everyone with health insurance.

What a dramatic change! I think what Schumpeter and Senior teach us is that we need to compose the folk songs for capitalism. Don't be ashamed to sing loudly and proudly about how much the market loves you. If you don't love it back, progress could leave you.

Baumol Cost Curse / Disease

Modern economies are wonderful things. If you really want to re-enact the middle ages, you should eat a healthy portion of your own soil or insects and develop scurvy. Not to mention let go of a few of your personal liberties.

Baumol's thesis concluded that while productivity was rising virtually across the board, the hidden cost was on things that respond little to capital-augmentation. His example is a live-orchestra. It took 250 people to perform a concert in 1750 and it still does today. Now, I have always laughed off this concept because compact disks and MP3s have made the marginal cost of listening to this performance virtually zero. But in 1993 Baumol resurrected his thesis by applying it to health care, and this was published in my favorite journal:

If you think about health care, you will start to realize that good health care is a function of personal relationships -- services. These relationships cannot be augmented by machines the same way that you can get more widgets from a worker by giving them a row of widget stamping machines and training them to monitor and maintain the machine. For pressure ulcers (bed sores), as one example, the patient needs to be clean, dry, and massaged on a routine basis (the sore develops local gangrenous material as a result of damp skin that has insufficient blood flow, a similar to trench foot anywhere there is contact between your body and the bed). All attempts to mechanize relief to this problem seem to fail.

In medicine you need highly skilled people to do tasks that are difficult to scale. Like someone who learns to play a violin, a proper nurse is a very talented artist. Because the increases in productivity here are small, the cost of the level of service rises in proportion to the opportunity costs.

Market advocates might respond that when the opportunity cost gets high enough, the incentive to find a capital-based solution will be so great as to create the needed innovation, but I am not particularly comfortable with necessary but not sufficient arguments. Similar to the market response to the energy crisis, the rising price of oil creates a necessary but not sufficient incentive to innovate alternative energy solutions.

I will continue to work on this idea and most likely post additional comments on it. I would love your thoughts!

Mill's Wage Fund and Interventionism

While Diana has been thinking about experts in the last year or so (see below), I have still been thinking about interventionism. Like her, I see this everywhere once I begin to think about it.

One of the great formations of this argument was by Ludwig von Mises. He took socialism seriously, and being from Vienna, he saw some of the many different forms socialism can take. Having sat in Cafe Central near the Ring Strasse in Vienna and sipped strong coffee, I can somewhat relate to the revolutionary ideas of Trotsky. However, being a native English speaker, I have always had a deep love for the debate waged in England and for too long under-appreciated the continent. I think that Mises looked to Mill for a strong defense of Socialism precisely because he knew the importance of the English tradition for the way economics was headed. After all, Mill was not in it for the revolutionary ideas, and both Mill and Mises took seriously the critique of capitalism where they found its critics to be intellectually honest.

Today, when I was making my history of thought test, I thought about how the intervention concept might relate to Mill's wages fund. Not to downplay Mises, but Mill's concept of the wages fund states that there is a certain amount of profit (revenue from capital) that can go to labor, a forerunner of the idea of labor's share of income. If the labor share of income is somewhat stable (an assumption that has empirical backing), then trying to raise the wage will cause unemployment. Now the cause of the unemployment is somewhat subtle. Mill didn't formulate the theory in a way modern economists can understand. That is to say he didn't use math. So, in an important way, we don't know exactly what he meant. However, if we follow my connection to the labor share of income, then the theory is not entirely useless as a pattern prediction.

This brings me to what Diana and I have been working on. Pattern predictions are statements about regularities in the world. They are not like neoclassical models because they are not expected to be replicated or to exist as controlled experiments. Misesian a priorism, in my opinion, is best categorized as pattern prediction. Mill had his own connections to a priorism. This has derogatorily been called the "Ricardian Vice" meaning the drastic oversimplification of economic models. When Mill is read this way, he is clearer. We raise the minimum wage (through unions) then that amount available for wages is distributed to the lucky recipients of the higher wage. The other portion of the labor employed loses this amount, but not in wages but by losing their jobs.

Now, I am going to be ahistorical. What do we learn from this example? How can employers pay more for labor than what it is worth under a competitive market. I think the answer to this is very simple. There must be some capital substitution (here is a modern example). Because wages rise the employer expects to see more productivity from that wage. The best way to do this is to increase capital investment. If wages are going to remain high for some time, the plant needs better machines which will increase productivity and drive down the number of labor hours needed to run the plant. We do need capital investment, but over-investment can be defined here as the rate which increases the unemployment rate.

This hardly redeems Mill, but it does suggest that Mill was on to something. Like Mises, he found a place where well intended interventions disrupted the rate of growth in capital and the cost of this acceleration in growth is passed onto those with the lowest skills in society. I find this much more intuitive and would love to test it to see if the intuition holds.

Pattern predictions state that all interventions create disruptions. Lipsey and Lanchaster still made the argument most clearly to their contemporary economists in the late 1960s. However, Mises made the point in a way that was meant to be understood by well-intended policy makers. Other examples include: Robert Higgs who talks about intervention when he says war help to grow the size of government through a ratchet effect. Buchanan talks about intervention coming directly as a result of not having a binding budget constraint on government. Al Roth applies the term repugnance to a wide category of interventions. My dissertation work tried to take this last piece and connect it to the story of the rise of economics in the last 300 years. I continue to find examples that confirm my reading. I still think that economics is the story of pointing out that interventions have consequences. Often, as in the case of Mill's wages fund, these consequences have high human costs among the least-well-off in society. It that lesson could be learned, I think most support for society wide interventions would erode.